After all, if they could predict the future, why did they design a game rather than get rich in the stock market?
https://ourworldindata.org/grapher/installed-solar-pv-capaci...
https://news.ycombinator.com/item?id=20782346
(Needless to say, I don't have a high opinion of current fusion start ups.)
And fission already had the financial fiasco of the first nuclear build out in the US.
Solar could have been predicted to get really cheap, if one had heroically projected its experience curve into the future. A lot of people chickened out doing that, even Greenpeace.
[0] Yes pollution from normal everyday fossil fuel power generation causes more damage and deaths in the aggregate but it's a diffuse hard to attribute kind of damage which makes it far less, for lack of a better way to say it, real to people. eg: Did living downwind of that unupgraded coal plant really kill them or is their death because of something else.
To borrow the meme about LISP [1], if nuclear was so good, it would've caught on at any point in the last 80 years. And it hasn't for some good reasons:
- It's too expensive. As an example, Hinkley Point-C in the UK has a super high negotiated rate that only lasts 35 years for a plant that still needs a lifetime of 60 to justify the investment so who knows how expensive the power will be for those last 25 years;
- It takes too long to build, typically ~20 years;
- Fuel processing (to create the fuel rods) creates a lot of harmful waste (eg UF6) and it is not a solved problem;
- Processing of spent fuel is not a solved problem. There's this idea that you just stick it in a barrel and leave it for a thousand years doesn't match the reality. It typically needs active cooling for possibly decades, for a start;
- Failure modes for a single plant dwarf any other alternative. The standard counterpoint is global warming from fossil fuels but that's for the entire fossil fuel ecosystem. The clean up for Fukushima, for example, will likely take over a century and cost over $1 trillion;
- The taxpayer bears the burden for these failures because, in the US for example, the industry self-insurance fund is a couple of orders magnitude too small to cover it;
- Nuclear requires liability limits well below any such failure mode to make it remotely economical (eg the Price-Anderson Act liability limits);
- The ecosystem of building, maintaining, repairing, servicing, monitoring and regulating nuclear power plants as well as all the supply chains (eg transporting and storing fuel) takes a long time to scale up.
I really wish I understood why nuclear has this cult-like following. It's completely irrational.
I believe it is largely because it is so unviable. If you can shut down any power supply argument with "but muh nuclear" then you can ensure that we keep burning coal and gas forever. And as a bonus, you get to pretend you're the smartest person in the room for having come up with the only good solution.
Because “cost” is not an argument against something that hasn’t been scaled. Nuclear was snuffed in the crib before any repeatable plants could be brought online.
If you looked at the cost of solar in 1995, using it to power anything like a house was a complete joke. It had to fall over an order of magnitude and leave behind a trail of solar company corpses to get to a point where you can break even after 10 years putting them on a house.
Solar and wind need storage to make a complete solution, so when it comes to climate change discussions, you just end up with two “irrational” groups arguing about the optimistic potential things that don’t actually exist at an affordable cost.
We tolerate all kinds of externalities from fossil fuel energy production and environmental degradation due to rare earth mining required to produce solar panels and wind turbines.
Nuclear energy is the only industry required to mitigate all possible externalities, and for that reason it is very expensive.
It’s disingenuous to claim nuclear power is a failure because it’s expensive to build without examining the reasons why that might be.
For proof it can work and is both safe and economical, witness the Chinese currently building 40 new nuclear plants. We could do that here, but choose not to. The only missing ingredient is political will, and resistance from ill-informed people such as yourself.
Is there a plant where the enviromental costs of building and decomissioning a plant and of extracting and enriching the fuel are completely mitigated?
I love this argument because it's never qualified. What regulations exactly make nuclear unaffordable? How many of them are safety related? Is safety not necessary?
It's just a hand-wavy argument for nebulous deregulation when all the evidence suggests that no industry is capable of self-regulation.
> ... rare earth mining required to produce solar panels ...
Solar panels don't require rare earths.
> Nuclear energy is the only industry required to mitigate all possible externalities
This is just objectively false. Fuel processing, spent fuel processing and storage and failure modes are absolutely not completely mitigated or even solved problems.
> For proof it can work and is both safe and economical, witness the Chinese currently building 40 new nuclear plants.
Well if 40 power plants are being built, you can't really say that their operational record is that they're "safe and economical" because they have no operational record, right? Now China has other nuclear power plants but you're not arguing with those.
China has several advantages no Western country has:
1. A standardized design so economies of scales with components;
2. State subsidies such as ultra-cheap finance; and
3. If people who manage and oversee such plants skimp on such maintenance they'll likely be executed.
Games like SimCity often only supports ultradense Manhattan style cities and centralized solar plants through distribution networks. That combination doesn't work with solar technology. It should be like "Set up offgrid solar incentives program($1.5B)" buttons that reduces power draw proportionate to numbers of standalone houses that resets when disaster hits the parts of the map it was used on.
Solar seems to be working just fine in Spain, a place where almost nobody lives in a stand alone house, or even a row house. You still put solar in the building, but then you have industrial solar in places wi5 little agricultural value, and wind turbines installed on top of hills. All of this could work just fine in any sim city I remember, as you are just dedicating space to power, which is fine.
The situation is slightly different for rooftop solar installations, which are equally prone to flying debris as concentrated farms but somewhat less likely to be affected by disasters that the house it is on can withstand. But that would be for off-grid purposes and not for the grid, and my original comment that solar generation only works for on-prem off-grid applies.
Plan it and it won't be a huge deal.
That said, I'm sure there are spots on this planet where solar isn't the best renewable option. Maybe small-scale wind power would be better where you are? On our boat we have both (and hydro for when we're sailing). Sadly these mechanical generators haven't yet received similar scale effects to their price as solar has.
Plus below a certain cost threshold, the risk of damage still means the overall cost is lower.
Finally, some much needed progress towards Sagan's apple pie.
Nuclear's problem, despite it's Champions assertion that it's now safe and properly engineered, is it's still inherently dirty and that dirt follows it through out it's life. It also probably has bloody mines, but every step on the path of using nuclear requires security and safety, and even when it's disposed, it still requires safety and security. It loses because no one trusts anyone along the entire value chain to not cut corners that leads to unsafe practices.
I think there was some ex-French colony government that demanded France disclose details of nuclear waste burial sites for "safety" reasons, which France rejected. Which means it happens. That kind of thing is why nuclear wastes are supposedly and actually so "dangerous" and small private owned generators get "complicated".
It's from 2022, so probably outdated.
> According to the US Department of Energy (DOE), about 12% of all silicon metal produced worldwide (also known as “metallurgical-grade silicon” or MGS) is turned into polysilicon for solar panel production. China produces about 70% of the world’s MGS and 77% of the world’s polysilicon. Converting silicon to polysilicon requires very high temperatures, and in China it’s coal that largely fuels these plants. Xinjiang—a region in China of abundant coal and low electricity prices—produces 45% of the world’s polysilicon.
> Reports indicate that some Xinjiang polysilicon plants have employed forced labor of Uyghurs, an intensely persecuted Muslim ethnic minority.
[0] https://blog.ucs.org/charlie-hoffs/mining-raw-materials-for-...
That doesn't change that you've made a total apples to oranges comparison by using the environmental cost of nuclear fuel to the cost of having a solar panel.
Both produce a finite amount of MWh for their lifetime. Both require mining. How much mining and human rights violations exist per unit of energy produced is the metric you're looking for.
That's complicated though. Easier to pretend like it's zero for one and non-zero for the other.
> It needs mines, but those are easily forgotten about once the blood money is washed through a few hands.
This sentence was commenting on solar panels.